Geographical influences upon physical activity participation: evidence of a ‘coastal effect’

Brief Report

Geographical influences upon physical activity participation: evidence of a ‘coastal effect’ Adrian Bauman

Abstract Objective: To examine the association

between geographical proximity to the coast and physical activity participation levels. Method: Using stratified random sampling, a telephone survey was carried out in 1994 with 1000 adults in each of the 16 health service regions in New South Wales (N=16178).Physical activity levels were measured through self-report of the frequency and duration of walking, moderate and vigorous activities in the two weeks preceding the survey. Logistic regression modelling was carried out to examine the association between physical activity and ‘coastal’location of residence, adjusting for age, sex, employment status, education level and country of birth. Results: After adjusting for other demographic factors, respondents who lived in a coastal postcode were 23% less likely to be classified as sedentary, 27% more likely to report levels of activity considered adequate for health, and 38% more likely to report high (vigorous) levels of physical activity than those who lived inland. Each of these associations was significant at the 0.05 level. Conclusions: Characteristics of the physical environment in coastal postcodes are related to physical activity participation. Implications: Physical environments may contribute to physical activity participation. Further efforts to conceptualise and measure these environmental influences is warranted. Public health efforts to promote physical activity should consider aspects of the physical environment as part of any intervention. (Aust N Z J Public Health 1999;23:322-4)

Submitted: August 1998 Revision Requested: January 1999 Accepted: April 1999

322

School of Community Medicine, University of New South Wales

Ben Smith Epidemiology Unit, South Western Sydney Area Health Service

Lyn Stoker Centre for Education and Information on Drugs and Alcohol, New South Wales

Bill Bellew Health Promotion Branch, New South Wales Health Department

Michael Booth Australian Centre for Health Promotion, Department of Public Health and Community Medicine, University of Sydney

T

he potential influence of the physical environment on physical activity behaviour is increasingly recognised,is2 yet the nature of this relationship is not well understood. It has been suggested environmental factors may enable or present barriers to physical activity participation, act as cues to exercise, reinforce regular activity, and maintain and possibly mediate the effects of educational campaigns about physical a ~ t i v i t y . ~ . ~ The potential for environmental interventions to promote activity are substantial. Physical environments usually possess fixed and enduring qualities and can affect a large number of people, and so are thought to be useful in developing interventions to increase physical activity parti~ipation.~ Emphasis on the environment’s contribution also comes from a social ecology perspective in health behaviour research. This considers health behaviours to be the product of a dynamic relationship betwen personal characteristics and facets ofthe social and physical en~ironment.~.’ Research to support the importance given to the physical environment in relation to activity patterns has been limited. Cross sectional studies have found a weak relationship between perceptions o f the local

environment and minutes of walking for exercise8 and no relationship with levels of vigorous a ~ t i v i t yLongitudinally, .~ perceptions of the quality ofthe neighbourhood environment have been found to predict adoption of vigorous activity by men over a two-year period.’O One study which applied an objective measure of the physical environment found proximity to pay exercise facilities to be positively related to vigorous activity.“ Interestingly, perceived convenience of facilities did not correlate with objectively measured proximity in this study. Even less is known about how environmental interventions can bring about increases in physical activity. A controlled study in a US Naval base found that a range of enhancements to exercise and health facilities at the site increased performance on fitness tests, but had no effect on leisure-time kilocalorie expenditure.I2 Other research has found that simple environmental manipulations, in the form of signs advocating the use the stairs instead of elevators, have increased stair use, but usually only in the short term.13*14 The need for more descriptive research which can indicate the contribution of physical environment variables to exercise in the

Correspondence to:

Professor Adrian Bauman, Epidemiology Unit, Liverpool Hospital, Liverpool 21 70 NSW. Fax: (02) 9828 6012; 6-mail: a.baumanOunsw.edu.au

AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH

1999 VOL. 23 NO. 3

Brief Report

Geographical influences upon physical activity participation

presence of other factors has been re~ognised,~ particularly where objective rather than subjective measures of the environment are used. This will give a clearer indication of the importance that such environmental characteristics might have in influencing physical activity participation. This paper aims to address this need by examining the relationship between proximity to the coast and various measures of physical activity using data from a telephone survey of 16178 people in New South Wales. Specifically, the association between coastal residence and levels of sedentariness, adequate energy expenditure and vigorous activity will be examined, adjusting for sex, age, ethnicity, education and employment status.

Methods Sample In 1994 a stratified random sample of households in New South Wales was taken using the electronic telephone directory. One thousand adult respondents were sought in each of the 16 health service regions in New South Wales. At each household contact telephone interviewers asked to speak to the person who was over 18 years of age and had the most recent birthday. Bilingual interviewers were available to conduct the survey in major languages. Measurement A comprehensive survey addressing major health risk factors was carried out. Physical activity levels were measured through self-rcport of the frequency and duration of walking, moderate and vigorous activities in the two weeks preceding the survey. These sclf-report measures have demonstrated reliability and validity.”,16A range of demographic data were collected, including

age, sex, education level, employment status, country of birth, language spoken at home and postcode of residence.

Data analysis Data were analysed using SPSS for Windows V6.0. Applying information about self-reported activity and weight, energy expenditure levels were calculated in terms of estimated kilocalories per week (kcal/wk). Three physical activity outcome variables were used.These were: (i) being totally sedentary, below 50 kcal/wk; (ii) being adequately active for health, greater than 800 kcallwk, and; (iii) vigorously active, greater than 1600 kcal/wk and at least one hour of vigorous activity. Location of residence was categorised as “coastal” if the individual’s postcode touched the NSW coastline, otherwise it was deemed to be “inland”. In order to test the hypothesis that coastal proximity was associated with physical activity, forced entry logistic regression was carried out. Modelling examined the effect of location of residence upon activity, after adjusting for age, sex, employment status, education level and country of birth

Results Respondents There were 16,178 telephone interviews completed, representing a response rate of 72.6%. The characteristics of respondents are included inTable 1. Detailed comparisons of respondent characteristics with those of the NSW population are given elsewhere.” Compared with the 1991 Census Profile of NSW residents, survey respondents were slightly more likely to be female, born in an English speaking country and over 40 years of age.

Table 1:Adjusted odds ratios and 95% confidence intervals for associations between geographic residence and other demographic variables and level of physical activity participation. Sedentary

Respondents %

Yo Adjusted OR (95% Cl)

.

Gender fn=16.1781 , Male Female

Adequate weekly energy expenditure Yo Adjusted OR (95% CI)

High level of activity % Adjusted OR (95% Cl)

42.4 57.6

12.4 13.0

1.02 (0.92. 1.14)

59.1 43.5

0.53 (0.49, 0.56)

22.9 12.8

0.49 (0.45. 0.54)

39.7 33.9 26.3

10.6 13.5 15.1

1.20 (1.07, 1.34) 1.28 (1.09, 1.50)

51.8 48.7 49.3

0.9 (0.84, 0.97) 0.9 ( 0.8, 1.00)

25.4 15.0 7.3

0.55 (0.50, 0.61) 0.37 (0.31. 0.44)

88.3 11.7

12.5 14.6

1.24 (1.07, 1.42)

51.2 41.7

0.63 (0.57, 0.70)

17.7 12.6

0.59 (0.51, 0.69)

18.4 44.5 37.1

20.0 11.9 10.4

0.57 (0.51, 0.65) 0.48 (0.42, 0.54)

42.1 50.2 53.8

1.43 (1.30, 1.57) 1.64 (1.48, 1.81)

6.8 17.5 21.8

1.97 (1.67, 2.32) 2.35 (1.99, 2.77)

Employmenr [n= 16,0931 Full or part time Retired or unable to work Home duties Unemployed Student

56.5 19.1 18.2 3.3 3.1

12.4 14.2 14.5 7.4 6.6

0.84 0.95 0.51 0.52

0.99) 1.10) 0.71) 0.76)

50.8 52.9 42.7 57.5 56.0

1.28 (1.14, 1.11 (1.01, 1.39 (1.16, 1.38 (1.14,

1.44) 1.23) 1.67) 1.67)

21.5 7.8 10.6 22.7 28.6

0.64 (0.52, 0.77) 0.86 (0.74, 1.OO) 1.11 (0.89, 1.38) 1.37 (1.11, 1.69)

Residence [n= 16.1 781 Inland Coastal

77.0 23.0

13.4 10.7

0.77 10.69, 0.87)

48.6 55.2

1.27 (1.18, 1.37)

16.4 19.6

1.38 (1.25, 1.52)

~.

A g e In= 16,1771 c 40 years 40-59 years 60 years and over ~.

Country of birth [n=16,178] English speaking Non-English speaking ___ Education [n=15,982] Some high school or less Completed high school Tertiary ~

.

(0.71, (0.82, (0.37, (0.36,

Note: (a) percentages do not add to 100 due to rounding e r m

1999 VOL. 23 NO. 3

AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH

323

Bauman et al.

Brief Report

Correlates of physical activity In relation to the central question in this paper, Table 1 shows that, after adjusting for other major demographic factors, those who lived in a coastal postcode were 23% less likely to be sedentary, 27% more likely to report adequate activity, and 38% more likely to report high levels of physical activity. The table also shows the other demographic factors significantly associated with all three of these physical activity variables. People less likely to be sedentary and more likely to have higher rates of adequate or high energy expenditure activity were: aged 40 years or less (except in comparison to those over 60 years for levels of adequate energy expenditure); born in an English speaking country; with a high school or tertiary level education, and; who were students compared to those in full or part time employment, Men had significantly higher rates of adequate activity than women.

Discussion We found that proximity to the coast was independently associated with higher levels of physical activity after adjusting for other important covariates - sex, age, country of birth, level of education and employment status. It was notable that this association was significant in the presence of covariates that are also strongly related to activity patterns. This descriptive finding lends support to previous claims that the physical environment may be an important influence upon physical activity participation. A strength of this investigation is that we were able to show, in support of previous research," that characteristics of the physical environment are related to activity regardless of individual environmental perceptions. The difficulties in obtaining valid measures of environmental perceptions have been recognised,s and there is also the risk that such perceptions may simply reflect underlying beliefs and attitudes that are the primary determinants ofphysical activity. This does not, however, discount the mediating role that perceptions may play'' and the development of valid measures of these factors is an important task. We have also found that objective environmental characteristics relate to sedentariness and inadequate physical activity, which adds to previous work which only investigated associations with vigorous activity." This suggests that environmental variables may play an important role in terms of the public health aspects of physical activity. While our ability to deduce causality is hindered by the crosssectional nature of this study, these findings can be used to build hypotheses about the environment-exercise relationship. The coastal environment may nurture mediating factors that are related to exercise participation, such as higher self-efficacy for physical activity, fewer perceived barriers and greater exercise enjoyment. Particular attributes of the coastal setting such as proximity to recreation spaces, facilities that can be used free ofcharge and aesthetic qualities may positively influence these mediating factor^.^,'^ More detailed population surveys, with sample sizes adequate to make comparisons across different geographical settings, would be valuable to quantify the strength of relationships between possible mediating factors, specific characteristics of the 324

environment and physical activity. Longitudinal studies incorporating measurement of these variables could indicate the role of environmental factors in influencing the adoption and maintenance of regular activity after adjustment for possible mediating factors. An alternative hypothesis that can be drawn from these findings is that the coastal location is simply the preferred living environment of people who place a higher value upon physical activity. Income may be a factor here, as higher income people may be both more likely to participate in regular physical activity and to live in coastal locations. Cultural norms regarding activity may also be interrelated with geographical location and have a subsequent impact upon behaviours.This process might be further illuminated through qualitative research. The dearth of knowledge about the link between the physical environment and physical activity is limiting progress in this field. We have shown a link between objective environmental characteristics and activity which lends further support to the need for better conceptualisation of environmental indicators in health promotion and more effective interventions in this important area of public health.

References I. King AC, Blair SN, Bild DE, et al. Determinants of physical activity and interventions in adults. Med Sci Spurts Exerc 1992; 24(6 Suppl): 221-36. 2. Marcus BH. Exercise behaviour and strategies for intervention. Res Q E.rerr Sport 1995; 66(4): 319-23. 3. Laitakari J. Vuori I , Oja I? Is long term maintenance of health-related physical activity possible? An analysts of concepts and evidence. M d t h Edur Rt.s 1996; l l ( 4 ) : 463-77. 4. King AC, Jeffrey RW, Fridinger F, et al. Environmental and policy approaches

5.

6. 7.

8.

9.

10 11

12 13 14

15

to cardiovascular disease prevention through physical activity: Issues and opportunities. Health Educ Q 1995; 22: 499-511. Sallis JF, Owen N. Ecological models. In: Glanz K, Lewis FM, Rimer BK, editors. Health behaviour und health education: theory, research and practice. 2nd ed. San Francisco: Jossey Bass, 1997. Stokols D. Establishing and maintaining healthy environments: towards a social ecology of health promotion. Am Psycho/ 1992; 47(1): 6-22. Green LW, Richard L, Potvin L. Ecological foundations of health promotion Am J Health Promot 1996; lO(4): 270-81. Hovell MF, Sallis JF, Hofstetter C R . Identifying correlates of walking for exercise: an epidemiologic prerequisite for physical activity promotion. Prc,\, Med 1989; 18: 856-66. Sallis JF, Hovell MF, Hofstetter CR. A multivariate study of the determinants of vigorous exercise in a community sample. Prev Med 1989; 18: 20-34. Sallis JF, Hovell MF, Hofstetter CR. Predictors of adoption and maintenance of vigorous physical activity in men and women. Prev Med 1992; 21: 237-5 I Sallis JF, Hovell MF, Hofstetter CR, et al. Distance between homes and exercise facilities related to frequency of exercise among San Diego residents. Public Healrh Rep 1990; 105(2): 179-85. Linenger JM, Chesson IICL( Nice DS. Physical fitness gains following simple environmental change. Am J Prev Med 1991; 7(5): 298-310. Brownell KD, Stunkard AJ, Albaum JM. Evaluation and modification of exercise patterns in the natural environment. A m JPsychiutry 1980; 137: 1540-5 Blamey A, Mutrie N, Aitchison T. Health promotion by encouraged use of stairs. Er Med J 1995; 31 1: 289-90. Booth M, Owen N, Bauman A, Gore C. Retest reliability of recall measures of leisure time physical activity in Australian adults. Int J Epidemiol 1996;

25: 153-9. 16 Booth ML, Owen N, Bauman A, Gore C. Relationship between a 14-day

recall measure of leisure time physical activity and a submaximal test of physical work capacity in a population sample of Australian adults. Res Q Exerc Sport 1996; 67: 221-7. 17 NSW Health Department. NSW Health Promotion Survqv 1994. Sydney: National Centre for Health Promotion, NSW Health Department, 1995. 18 Holman CDJ. Measuring the occurrence of health-promoting interactions with the environment. Aus/ N Z J Public Healrh 1997; 21(4): 360-4. 19 Corti B, Donovan RJ, Holman CDJ. Factors influencing the use of physical activity facilities: Results from qualitative research. Health Promot J Aust 1997; 7 : 16-21.

AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH

1999 VOL. 23 NO. 3